linux-clk.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Jun Nie <jun.nie@linaro.org>
To: Philipp Zabel <p.zabel@pengutronix.de>
Cc: Michael Turquette <mturquette@baylibre.com>,
	sboyd@kernel.org, Rob Herring <robh+dt@kernel.org>,
	Mark Rutland <mark.rutland@arm.com>,
	Wei Xu <xuwei5@hisilicon.com>,
	opensource@jilayne.com, swinslow@gmail.com, allison@lohutok.net,
	yuehaibing@huawei.com, tglx@linutronix.de,
	linux-clk@vger.kernel.org, devicetree@vger.kernel.org,
	linux-arm-kernel@lists.infradead.org, xuejiancheng@hisilicon.com
Subject: Re: [PATCH 2/3] reset: hisilicon: Extend reset operation type
Date: Tue, 3 Dec 2019 11:53:50 +0800	[thread overview]
Message-ID: <CABymUCNDZSH+mB9TyyUBwgRu-qTRbgUv89va2HuBs4VeJWn6uA@mail.gmail.com> (raw)
In-Reply-To: <449968d8f085a1d1fcf4018bb8efe454fa35b3e3.camel@pengutronix.de>

Philipp Zabel <p.zabel@pengutronix.de> 于2019年12月3日周二 上午1:04写道:
>
> On Mon, 2019-12-02 at 22:45 +0800, Jun Nie wrote:
> > Extend reset operations to support combination of three type flags:
> > ASSERT/DEASSERT SET/CLEAR POLL.
> >
> > Signed-off-by: Jun Nie <jun.nie@linaro.org>
> > ---
> >  drivers/clk/hisilicon/reset.c | 58 ++++++++++++++++++++++++++++++++---
> >  1 file changed, 53 insertions(+), 5 deletions(-)
> >
> > diff --git a/drivers/clk/hisilicon/reset.c b/drivers/clk/hisilicon/reset.c
> > index 93cee17db8b1..de7d186b0894 100644
> > --- a/drivers/clk/hisilicon/reset.c
> > +++ b/drivers/clk/hisilicon/reset.c
> > @@ -2,20 +2,25 @@
> >  /*
> >   * Hisilicon Reset Controller Driver
> >   *
> > - * Copyright (c) 2015-2016 HiSilicon Technologies Co., Ltd.
> > + * Copyright (c) 2015-2019 HiSilicon Technologies Co., Ltd.
> >   */
> >
> >  #include <linux/io.h>
> > +#include <linux/iopoll.h>
> >  #include <linux/of_address.h>
> >  #include <linux/platform_device.h>
> >  #include <linux/reset-controller.h>
> >  #include <linux/slab.h>
> >  #include <linux/spinlock.h>
> > +
> > +#include <dt-bindings/reset/hisilicon-resets.h>
> >  #include "reset.h"
> >
> >  #define      HISI_RESET_BIT_MASK     0x1f
> >  #define      HISI_RESET_OFFSET_SHIFT 8
> >  #define      HISI_RESET_OFFSET_MASK  0xffff00
> > +#define      HISI_RESET_FLAG_SHIFT   24
> > +#define      HISI_RESET_FLAG_MASK    0xff000000
> >
> >  struct hisi_reset_controller {
> >       spinlock_t      lock;
> > @@ -30,14 +35,17 @@ struct hisi_reset_controller {
> >  static int hisi_reset_of_xlate(struct reset_controller_dev *rcdev,
> >                       const struct of_phandle_args *reset_spec)
> >  {
> > +     unsigned long flags;
> >       u32 offset;
> >       u8 bit;
> >
> > +     flags = (reset_spec->args[2] << HISI_RESET_FLAG_SHIFT)
> > +             & HISI_RESET_FLAG_MASK;
> >       offset = (reset_spec->args[0] << HISI_RESET_OFFSET_SHIFT)
> >               & HISI_RESET_OFFSET_MASK;
> >       bit = reset_spec->args[1] & HISI_RESET_BIT_MASK;
> >
> > -     return (offset | bit);
> > +     return (flags | offset | bit);
> >  }
> >
> >  static int hisi_reset_assert(struct reset_controller_dev *rcdev,
> > @@ -48,13 +56,33 @@ static int hisi_reset_assert(struct reset_controller_dev *rcdev,
> >       u32 offset, reg;
> >       u8 bit;
> >
> > +     flags = (id & HISI_RESET_FLAG_MASK) >> HISI_RESET_FLAG_SHIFT;
> > +     if (flags & HISI_ASSERT_NONE)
> > +             return -ENOTSUPP; /* assert not supported for this reset */
>
> As long as .assert and .deassert are the only implemented operations for
> this reset controller, this does not make any sense to me. Are there
> resets that can only be deasserted?

Some reset is asserted on power on automatically. So only .deassert is needed.
>
> > +
> >       offset = (id & HISI_RESET_OFFSET_MASK) >> HISI_RESET_OFFSET_SHIFT;
> >       bit = id & HISI_RESET_BIT_MASK;
> >
> > +     pr_devel("%s %s to %s 0x%x:bit[%d]\n", __func__,
> > +             flags & HISI_ASSERT_POLL ? "poll" : "",
> > +             flags & HISI_ASSERT_SET ? "set":"clear", offset, bit);
> > +
> > +     if (flags & HISI_ASSERT_POLL) {
>
> Since HISI_ASSERT_POLL is 0, this is always false.

Will fix the wrong value definition in next version patch. The same to
below comments.
>
> > +             if (flags & HISI_ASSERT_SET)
> > +                     return readl_poll_timeout(rstc->membase + offset,
> > +                                               reg, reg & BIT(bit), 0, 5000);
> > +             else
> > +                     return readl_poll_timeout(rstc->membase + offset,
> > +                                               reg, !(reg & BIT(bit)),
> > +                                               0, 5000);
>
> And this is effectively dead code. Do you really have hardware that
> asserts or asserts a reset line in reaction to a read access?
>
> Should HISI_ASSERT_POLL and HISI_DEASSERT_POLL be mutually exclusive?
>
> > +     }
> > +
> >       spin_lock_irqsave(&rstc->lock, flags);
> >
> >       reg = readl(rstc->membase + offset);
> > -     writel(reg | BIT(bit), rstc->membase + offset);
> > +     /* Default is setting to assert for no flag case. */
> > +     reg = (flags & HISI_ASSERT_CLEAR) ? reg & ~BIT(bit) : reg | BIT(bit);
>
> Consider moving this into a helper function with a boolean set/clear
> parameter.

Will do.
>
> As long as HISI_ASSERT_CLEAR is 0, the first branch is dead code.
>
> > +     writel(reg, rstc->membase + offset);
> >
> >       spin_unlock_irqrestore(&rstc->lock, flags);
> >
> > @@ -69,13 +97,33 @@ static int hisi_reset_deassert(struct reset_controller_dev *rcdev,
> >       u32 offset, reg;
> >       u8 bit;
> >
> > +     flags = (id & HISI_RESET_FLAG_MASK) >> HISI_RESET_FLAG_SHIFT;
> > +     if (flags & HISI_DEASSERT_NONE)
> > +             return -ENOTSUPP; /* deassert not supported for this reset */
> > +
>
> Are there resets that can only ever be asserted?

I do not know yet. Only extend this driver with all combination in logic.
>
> >       offset = (id & HISI_RESET_OFFSET_MASK) >> HISI_RESET_OFFSET_SHIFT;
> >       bit = id & HISI_RESET_BIT_MASK;
> >
> > +     pr_devel("%s %s to %s 0x%x:bit[%d]\n", __func__,
> > +             flags & HISI_DEASSERT_POLL ? "poll" : "",
> > +             flags & HISI_DEASSERT_SET ? "clear":"set", offset, bit);
> > +
> > +     if (flags & HISI_DEASSERT_POLL) {
> > +             if (flags & HISI_DEASSERT_SET)
> > +                     return readl_poll_timeout(rstc->membase + offset,
> > +                                               reg, reg & BIT(bit), 0, 5000);
> > +             else
> > +                     return readl_poll_timeout(rstc->membase + offset,
> > +                                               reg, !(reg & BIT(bit)),
> > +                                               0, 5000);
>
> See above, this code currently can never be reached.
>
> > +     }
> > +
> >       spin_lock_irqsave(&rstc->lock, flags);
> >
> >       reg = readl(rstc->membase + offset);
> > -     writel(reg & ~BIT(bit), rstc->membase + offset);
> > +     /* Default is clearing to deasseart for no flag case. */
> > +     reg = (flags & HISI_DEASSERT_SET) ? reg | BIT(bit) : reg & ~BIT(bit);
>
> Same as above, the read-modify-write for set/clear could be split out
> into a helper.
>
> > +     writel(reg, rstc->membase + offset);
> >
> >       spin_unlock_irqrestore(&rstc->lock, flags);
> >
> > @@ -103,7 +151,7 @@ struct hisi_reset_controller *hisi_reset_init(struct platform_device *pdev)
> >       rstc->rcdev.owner = THIS_MODULE;
> >       rstc->rcdev.ops = &hisi_reset_ops;
> >       rstc->rcdev.of_node = pdev->dev.of_node;
> > -     rstc->rcdev.of_reset_n_cells = 2;
> > +     rstc->rcdev.of_reset_n_cells = 3;
>
> This breaks current device trees (before patch 3). You can make sure
> device trees with #reset-cells = <2> keep working by parsing the #reset-
> cells and setting this value to 2 for old DTs.

All current dts file are modified accordingly. But existing dtb binary support
is an issue. Do you have any suggestion?

>
> >       rstc->rcdev.of_xlate = hisi_reset_of_xlate;
> >       reset_controller_register(&rstc->rcdev);
>
> regards
> Philipp
>

  reply	other threads:[~2019-12-03  3:54 UTC|newest]

Thread overview: 11+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-12-02 14:45 [PATCH 0/3] Extend Hisilicon reset type Jun Nie
2019-12-02 14:45 ` [PATCH 1/3] dt-bindings: clock: Update Hisilicon reset doc Jun Nie
2019-12-02 17:04   ` Philipp Zabel
2019-12-03  3:11     ` Jun Nie
2019-12-03 14:12       ` Philipp Zabel
2019-12-02 14:45 ` [PATCH 2/3] reset: hisilicon: Extend reset operation type Jun Nie
2019-12-02 17:04   ` Philipp Zabel
2019-12-03  3:53     ` Jun Nie [this message]
2019-12-03 14:15       ` Philipp Zabel
2019-12-04  6:09         ` Jun Nie
2019-12-02 14:45 ` [PATCH 3/3] ARM: dts: Update reset for hi3519 and hi3798cv200 Jun Nie

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=CABymUCNDZSH+mB9TyyUBwgRu-qTRbgUv89va2HuBs4VeJWn6uA@mail.gmail.com \
    --to=jun.nie@linaro.org \
    --cc=allison@lohutok.net \
    --cc=devicetree@vger.kernel.org \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-clk@vger.kernel.org \
    --cc=mark.rutland@arm.com \
    --cc=mturquette@baylibre.com \
    --cc=opensource@jilayne.com \
    --cc=p.zabel@pengutronix.de \
    --cc=robh+dt@kernel.org \
    --cc=sboyd@kernel.org \
    --cc=swinslow@gmail.com \
    --cc=tglx@linutronix.de \
    --cc=xuejiancheng@hisilicon.com \
    --cc=xuwei5@hisilicon.com \
    --cc=yuehaibing@huawei.com \
    --subject='Re: [PATCH 2/3] reset: hisilicon: Extend reset operation type' \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).